Consolidate scattered transcript formatting functions into single
cohesive class encapsulating all transcript processing logic
for better maintainability and clearer separation of concerns.
Add transcript cleaning step to remove spurious recognition artifacts
like randomly predicted "Vap'n'Roll Thierry" phrases that appear
without corresponding audio, improving transcript quality
by filtering model hallucinations.
When declaring scopes with our OIDC provider, they require us to prefix
each scope with our application name. This is to prevent reserving generic
scopes like rooms:list for only our app, as they manage a large federation.
I’m proposing a workaround where, if a resource server prefix is detected in
the scope, it’s stripped out. This solution is simple and sufficient
in my opinion.
Since the scopes are defined in the database, I don’t want to update
them directly. Additionally, each self-hosted instance may have a different
application name, so the prefix should be configurable via a Django setting.
While creating a meeting on mobile, the dialog was opening,
and when its width exceeds the mobile width, users are unable
to close the dialog.
Prevent the dialog opening on mobile as a hot fix.
Adjust z-index values to restore button interactivity broken by previous
z-index changes in commit 53e68b7, ensuring fullscreen warning dismiss
controls remain accessible to users instead of being blocked
by overlay layering.
Manually update libexpat to 1.6.53-r0 in Alpine 3.21.3 base image
to address CVE-2025-64720, CVE-2025-65018,
CVE-2025-66293 high-severity vulnerability until newer Alpine base image
becomes available, ensuring Trivy security scans pass.
Create initial resource server integration documentation based on existing
service account documentation structure to help developers understand
authentication flow and implementation requirements for external services
consuming Meet's protected resources.
Upgrade django-lasuite to v0.0.19 to benefit from the latest resource server
authentication backend. Thanks @qbey for your work. For my needs, @qbey
refactored the class in #46 on django-lasuite.
Integrate ResourceServerAuthentication in the relevant viewset. The integration
is straightforward since most heavy lifting was done in the external-api viewset
when introducing the service account.
Slightly modify the existing service account authentication backend to defer to
ResourceServerAuthentication if a token is not recognized.
Override user provisioning behavior in ResourceServerBackend: now, a user is
automatically created if missing, based on the 'sub' claim (email is not yet
present in the introspection response). Note: shared/common implementation
currently only retrieves users, failing if the user does not exist.
Define Docker network enabling external service providers to share Keycloak
instance with local development stack, supporting OIDC authentication flow
where services obtain tokens from shared Keycloak then pass to Meet
for introspection and validation.
Prepares Meet infrastructure for multi-service authentication architecture
though external service provider Docker Compose integration changes remain
in separate repository.
A manually constructed authentication URL didn’t match the actual endpoint
address, causing the Django backend to issue a 301 redirect to the correct URL.
This wasn’t a problem for regular users at first, but once a client integrating
through a virtual browser came on board, it became significant. The 301 redirect
was disrupting the virtual browser’s cookie/cache system, which in turn broke
the authentication flow.
This change aims to resolve the issue, although it’s not yet certain that
it will fully address their problem.
- add admin action to retry a recording notification to external services
- log more Celery tasks' parameters
- add multilingual support for real-time subtitles
- update backend dependencies
Mark unnecessary decorative icons as aria-hidden following feedback
from @cyberbaloo to eliminate redundant screen reader announcements
that create noisy and annoying experience for users relying on
assistive technologies.
Regenerate backend translation files to include missing translations for newly
added translatable strings in recent code changes, ensuring complete
internationalization coverage across all supported languages.
django-lasuite 0.1.16 changed the user update mechanism from .update()
to .save(), which triggers Django's constraint validation. This causes
an additional SELECT query to verify 'sub' field uniqueness on every
user update, despite 'sub' being immutable in our auth flow.
This commit update the test to make them pass again.
Restrict pylint version to 3.x in renovate configuration because
pylint-django 2.6.1 requires pylint<4, preventing automatic upgrades
to pylint 4.x that would create unresolvable dependency conflicts
until pylint-django releases compatible version.
Update brotli compression library to version 1.2.0 addressing
CVE-2025-6176 security vulnerability to maintain secure
compression functionality and pass security scans.
Update Django from previous version to 5.2.8 addressing CVE-2025-64459
and CVE-2025-64458 security vulnerabilities to maintain secure
application infrastructure and pass security audits.
Introduce new variables for the docker registry where to push docker images on forks:
- DOCKER_CONTAINER_REGISTRY_HOSTNAME for the docker registry hostname, with default value "docker.io"
- DOCKER_CONTAINER_REGISTRY_NAMESPACE for the docker registry namespace, with default value "lasuite"
We will introduce in the next commits the compose set-up that also
require examples values/config files. Thus, re-organize the kube ones
to dedicated folder, to make the files organisation extensible.
This release adds support for injecting custom Kubernetes
resources through the extraManifests parameter.
New features:
- Add extraManifests support for deploying custom resources
- Support multiple input formats (list, map, raw YAML strings)
- Enable Helm template variables in injected manifests
Add ability to inject custom Kubernetes manifests through the
values.yaml file. This allows users to deploy additional
resources (Deployments, Services, ConfigMaps, etc.) without
modifying the chart templates.
The template supports multiple input formats: list of objects,
map of named objects, and raw YAML strings, providing maximum
flexibility for users.
- Create templates/extra-objects.yaml with flexible rendering
- Add extraManifests parameter in values.yaml with documentation
- Support Helm template variables in injected manifests
- Handle list, map, and string YAML formats automatically
Add dynamic configuration for Deepgram STT via environment variables,
enabling multilingual real-time subtitles with automatic language
detection.
Changes:
- Add DEEPGRAM_STT_* environment variables pattern for configuration
- Implement _build_deepgram_stt_kwargs() to dynamically build STT
parameters from environment variables
- Add whitelist of supported parameters (model, language) for LiveKit
Deepgram plugin
- Log warnings for unsupported parameters (diarize, smart_format, etc)
- Set default configuration: model=nova-3, language=multi
- Document supported parameters in Helm values.yaml
Configuration:
- DEEPGRAM_STT_MODEL: Deepgram model (default: nova-3)
- DEEPGRAM_STT_LANGUAGE: Language or 'multi' for automatic detection
of 10 languages (en, es, fr, de, hi, ru, pt, ja, it, nl)
Note: Advanced features like diarization and smart_format are not
supported by the LiveKit Deepgram plugin in streaming mode.
Add detailed logging for owner ID, recording metadata, and
processing context in transcription tasks to improve debugging
capabilities.
It was especially important to get the created document id,
so when having trouble with the docs API, I could share
with them the newly created documents being impacted.
Replace narrow HTTPError handling with broad RequestException
catch to prevent crashes from network failures (ConnectionError),
timeouts (30s exceeded), SSL/TLS errors, and other request failures
that previously caused unhandled exceptions.
Ensures consistent False return and proper logging for all network-related
failures instead of crashing application when summary service
communication encounters infrastructure issues beyond HTTP errors.
While helping users, it was such a pain to determine quickly which recording
was indeed a transcription or a video recording.
Added the column to help me, and support team.
The recording / transcription is the most unstable part of the project.
Enable administrators to manually retrigger external service notifications
from Django admin for failed or missed notification scenarios,
providing operational control over notification delivery.
Remove 1000 z-index from screenshare warning that was
causing conflicts with reaction menu and reaction displays,
retaining only necessary layering to hide participant
metadata underneath.
Persist user preference choices across sessions using localStorage
following notification store pattern, eliminating need to reconfigure
disabled features on every meeting join and respecting user's
long-term preference decisions.
Adopt unified switch component pattern for notification preferences to
enable future addition of descriptive text per notification type,
improving consistency and providing clearer explanation capability
for notification behaviors.
Allow users to opt-out of idle participant disconnection despite
default enforcement, trusting power users who modify this setting
won't forget to disconnect, though accepting risk they may block
maintenance configuration updates.
Add margin between switch description text and toggle button to
improve visual breathing room and prevent text from appearing
cramped against interactive control element.
Introduce pop-in alerting participants of automatic 2-minute idle
disconnect to enable LiveKit node configuration updates during
maintenance windows, preventing forgotten tabs from blocking
overnight production updates following patterns
from proprietary videoconference solutions.
Introduce new narrow-width alert dialog variant to improve
readability of short messages by preventing excessively
long line lengths that occur when brief alerts use
standard dialog widths.
Expose idle disconnect timeout as configurable parameter accepting None value
to disable feature entirely, providing emergency killswitch for buggy behavior
without redeployment, following other frontend configuration patterns.
Restore correct task_args ordering in metadata manager after commit f0939b6f
added sender argument to Celery signals for transcription task scoping,
unexpectedly shifting positional arguments and breaking metadata creation.
Issue went undetected due to missing staging analytics deployment, silently
losing production observability on microservice without blocking transcription
job execution, highlighting need for staging analytics activation.
Posthog-cli version wasn't pinned.
Please check issue #39846, which describe our issue, starting
0.5.0, the cli needs an API token and a Project ID.
Pin to the last stable version we used 0.4.8, and wait a bit
they already released a 0.5.1 that mitigate some of the breaking
change.
I would wait the 0.5.x to be stable and battle tested by other
developpers before switching.
Also as I consider switching the Error tracking to sentry.
Fixes "Invalid LiveKit token" errors caused by field mismatch between
token generation and authentication lookup.
Previously:
- generate_token() used user.sub as token identity
- LiveKitTokenAuthentication tried to retrieve user via user.id field
- This failed when sub was not a UUID (e.g., from LemonLDAP OIDC provider)
Now:
- generate_token() continues using user.sub (canonical OIDC identifier)
- LiveKitTokenAuthentication correctly looks up by sub field
- Both sides now consistently use the same field
This ensures compatibility with all RFC 7519-compliant OIDC providers,
regardless of their sub claim format.
Without explicit commands in values.yaml,
celeryTranscribe and celerySummarize pods
were using the Dockerfile's default CMD (uvicorn),
which started the REST API instead of Celery workers.
This fix adds default commands to values.yaml for both services,
ensuring they run as Celery workers processing their respective
queues (transcribe-queue and summarize-queue).
Add ability to use response_format in call function in order to
have better result with albert-large model
Use reponse_format for next steps and plan generation
